AI Paper Atlas

Research map / Time series / Long-term forecasting

TSFAdv: Frequency-Guided Black-Box Adversarial Attacks on Time Series Forecasting

Qizhuo Han, Xiangrui Cai, Sihan Xu, Ying Zhang, Zheli Liu

ICML 2026 · Time series › Long-term forecasting · approach: Lightweight long-term forecasters

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add profiled amplitude-phase sensitivity weights to frequency-domain NES (V3 vs V1) (ablation)TSFAdv · 5 datasets averageDTW: 32.157 -> 43.940 (+37.0% DTW, +27.6% SME)Improved
use random weights instead of profiled sensitivity weights (V2 vs V1) (ablation)frequency-domain NES · 5 datasets averageDTW: 32.157 -> 16.708Hurt
move NES from time domain to frequency-domain parameterization alone (ablation)NES · 5 datasets averageDTW: 36.32 -> 32.157 (11.5% reduction in DTW)Hurt
use TSFAdv instead of NES for the attackDLinear · WeatherDTW: 27.565 -> 77.161 (179.92%)Improved
use TSFAdv instead of NES for the attackTimesNet · WeatherDTW: 203.5007 -> 206.1333 (1.29%)No clear effect
use TSFAdv instead of NES for the attackDLinear · ETTm1DTW: 13.9518 -> 14.5333 (4.16%)Mixed
double perturbation budget epsilon from 0.1 to 0.2 (ablation)TSFAdv on TimeMixer · ETT, 200 queriesDTW: 150% DTW improvement (vs. 100% for NES)Improved
apply randomized smoothing as defenseiTransformer under TSFAdv · ETT, 5 datasetsSME: 0.193 -> 0.266Hurt
apply magnitude pruning as defenseiTransformer under TSFAdv · 5 datasets averageDTW: 57.047 -> 58.118No clear effect
apply Gaussian-augmentation adversarial training as defensePatchTST under TSFAdv · 5 datasets averageDTW: 38.355 -> 48.819Hurt
apply adversarial training as defenseiTransformer under TSFAdv · 5 datasets averageDTW: 57.047 -> 52.416Improved
use sensitivity profile from a different ETT dataset (h2->h1 vs h1->h1) (ablation)TSFAdv · ETT domainDTW: 34.169 vs. 34.190No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy